A Joint Venture of Ab Initio Molecular Dynamics, Coupled Cluster Electronic Structure Methods, and Liquid-State Theory to Compute Accurate Isotropic Hyperfine Constants of Nitroxide Probes in Water

We have performed a large-scale evaluation of current computational methods, including conventional small-molecule force fields, semiempirical, density functional, ab initio electronic structure methods, and current machine learning (ML) techniques to evaluate relative single-point energies. Using up to 10 local minima geometries across ~700 molecules, each optimized by B3LYP-D3BJ with single-point DLPNO-CCSD(T) triple-zeta energies, we consider over 6,500 single points to compare the correlation between different methods for both relative energies and ordered rankings of minima. We find promise from current ML methods and recommend methods at each tier of the accuracy-time tradeoff, particularly the recent GFN2 semiempirical method, the B97-3c density functional approximation, and RI-MP2 for accurate conformer energies. The ANI family of ML methods shows promise, particularly the ANI-1ccx variant trained in part on coupled-cluster energies. Multiple methods suggest continued improvements should be expected in both performance and accuracy.

Download Full-text

Assessing Conformer Energies using Electronic Structure and Machine Learning Methods

10.26434/chemrxiv.11920914.v2 ◽

2020 ◽

Author(s):

Dakota Folmsbee ◽

Geoffrey Hutchison

Keyword(s):

Machine Learning ◽

Electronic Structure ◽

Density Functional ◽

Large Scale ◽

Single Point ◽

Semiempirical Method ◽

Coupled Cluster ◽

Scale Evaluation ◽

Machine Learning Methods ◽

Electronic Structure Methods

We have performed a large-scale evaluation of current computational methods, including conventional small-molecule force fields, semiempirical, density functional, ab initio electronic structure methods, and current machine learning (ML) techniques to evaluate relative single-point energies. Using up to 10 local minima geometries across ~700 molecules, each optimized by B3LYP-D3BJ with single-point DLPNO-CCSD(T) triple-zeta energies, we consider over 6,500 single points to compare the correlation between different methods for both relative energies and ordered rankings of minima. We find promise from current ML methods and recommend methods at each tier of the accuracy-time tradeoff, particularly the recent GFN2 semiempirical method, the B97-3c density functional approximation, and RI-MP2 for accurate conformer energies. The ANI family of ML methods shows promise, particularly the ANI-1ccx variant trained in part on coupled-cluster energies. Multiple methods suggest continued improvements should be expected in both performance and accuracy.

Download Full-text

Geometry of OH⋯O interactions in the liquid state of linear alcohols from ab initio molecular dynamics simulations

Physical Chemistry Chemical Physics ◽

10.1039/d0cp00435a ◽

2020 ◽

Vol 22 (12) ◽

pp. 6690-6697 ◽

Cited By ~ 3

Author(s):

Aman Jindal ◽

Sukumaran Vasudevan

Keyword(s):

Molecular Dynamics ◽

Hydrogen Bonding ◽

Ab Initio ◽

Molecular Dynamics Simulations ◽

Liquid State ◽

Crystalline State ◽

Md Simulations ◽

Ab Initio Molecular Dynamics ◽

Linear Alcohols ◽

Dynamics Simulations

Hydrogen bonding OH···O geometries in the liquid state of linear alcohols, derived from ab initio MD simulations, show no change from methanol to pentanol, in contrast to that observed in their crystalline state.

Download Full-text